since I did not found sufficient information in other threads I have to adress to another problem concerning the detection of the shortest route in my polyline network.
One of the routes between two points shows a uncommon behaviour. The route itself is flawless but the network analyst did not deliver the shortest route in this network. At the end of the calculation the route takes a small turn off and from that point to the end point it measures 48km. Without that it would be 47km. I can route error-free trough the 47km segments by including EVERY vertex while measuring with the "ruler measuring tool". Same for the 48km Segments...
Each others polylines connections and also the point feature connections with the polyline vertices are all topologically correct and there are no elevations or other restrictions in this network. Has anybody an idea?
What network dataset are you using? Is your data small enough to zip up and post here? If so, please post it along with a screen shot or MXD that has where your start and destination stops are located.
Otherwise, what type of connectivity policy are you using (any vertex or end point)? Also, you mention that there are no "other restrictions". Just to clarify, this means you have no turn restrictions, oneway restrictions, or anything like that? Do you have hierarchy enabled? If so (or if not), try the other setting and see if you get different results.
To do a little debugging, you could try to use the Network Identify tool (on the Network Analyst toolbar), to step along the route and make sure that the network has the connectivity you expect build into it for that area.
All of that is a good start. We'll work it out until we figure out what went wrong.